| |
|
General
Where can I get additional information about
neural networks?
There is a good introductory book written by Kevin Gurney and available
online at: http://www.shef.ac.uk/psychology/gurney/notes/index.html
You can also try Dr. Leslie Smith’s brief online introduction
to neural networks packed with pictures and examples at: http://www.cs.stir.ac.uk/~lss/NNIntro/InvSlides.html.
A good introductory book for managers and business analysts is:
Bigus, J.P. (1996), Data Mining with Neural Networks: Solving Business
Problems--from Application Development to Decision Support, NY:
McGraw-Hill.
For engineers and technically-minded people we’d recommend
to start with: Fausett, L. (1994), Fundamentals of Neural Networks:
Architectures, Algorithms, and Applications, Englewood Cliffs, NJ:
Prentice Hall.
For financial specialists, bankers and traders we recommend starting
with: E. Michael Azoff (1994). Neural Network Time Series: Forecasting
of Financial Markets NY: John Wiley and Sons, Inc.
How could I improve things to get better
forecasting?
You have two ways to improve results:
1) improve you input data (for more information please read Preparing
Data Sets section in Advanced Issues chapter)
2) improve network topology selection and network training (for
more information please read Selecting Network Topology and Training
Network sections in Advanced Issues chapter).
When neural networks are a bad choice for my forecasting?
Neural networks cannot create or digest the information that is
not contained in your data. To properly train a neural network you
need to have a lot of data. You data should contain input parameters
(signals, attributes, correlated values) that affect the target
value. Change of input parameters should lead to change of target
one.
So, if you have small amount of historical data or if you do not
know, which parameters influence your target value, better use some
other forecasting method.
In addition, there exist some problems that in principle cannot
be solved by neural networks. Do not use neural networks (as well
as other numerical methods) for problems like:
predicting random or pseudo-random numbers, like lottery
numbers
forecasting cash flow, volumes of sales, etc. if your business
isn’t stable and your market situation often changes dramatically.
any problem where historical data have no use due to unbiased,
rapid and significant changes in the problem environment.
 top
|
| |
|
Data Analysis and Preprocessing
How much historical data do I need?
You definitely need to have more records in the training subset
than the total number of input columns.
The number of records needed for training depends on the complexity
of your problem and amount of noise in your data. There are no exact
rules. Typically, it’s recommended to have at least 10 times
as many records for training as input columns.
This may not be enough for problems with subtle and complex dependencies
in data. Try to add more data if your network has poor results.
What is a categorical column?
Each value of a categorical column represents a certain category.
For example, categorical is a column that contains only “Male”
or “Female” as its values. Typically, the number of different
values in a categorical column is much less than the number of records.
Categorical data should be encoded in a special way to be suitable
for a neural network.
You may manually mark a column as categorical in Expert Mode (using
Details button at Data Analysis Progress step). This feature may
be beneficial for some cases. For example, your data has a column
“Model” that has values “1”, “2”,
“3”. By default, this column will be considered as a
numeric, but it will be more beneficial to encode it as a categorical
one.
Why Forecaster XL ignored some rows and columns?
This may happen if some of your columns or rows are unsuitable for neural
network. For example, text or data/time data cannot be processed by neural
network. Also, some your rows may have missing or invalid data; such rows
will be ignored.
To see which columns and rows were ignored look into Data
Preprocessing Report.
top
|
| |
|
Network Preparation
What is network training?
Network training means adjusting neural network weights. During
training the network analyzes the data you have provided and changes
weights between network units to reflect dependencies found in your
data.
What training algorithm Forecaster XL uses?
Forecaster XL uses constructive algorithm to train network and select the
network topology. This constructive algorithm is developed by Alyuda's
Research Group and is capable of automatic selection and tuning of training
parameters and network topology.
How Forecaster XL determines neural network topology
suitable for my problem?
See What training algorithm Forecaster XL uses?
What stopping conditions should I specify to improve
forecasting quality?
As the first step we recommend you using default settings that means your
network is trained until error reduction is no longer possible. If forecasting
error is still unacceptably high we recommend reducing MSE value, reducing the
error change value and increasing number of iterations.
Why I cannot see MSE and absolute error in the
network training report?
When your target column is not numeric, it is hard to define unambiguously
what the absolute error is. For such cases it is better to use correct
classification rate to let you know what percentage of data was recognized
correctly.
What is “error change” in stopping conditions?
Error change specifies the error change during several last iterations. This
parameter is useful for detection of situations when each new iteration has
almost no influence on error and thus the network cannot further improve its
performance and training should be stopped to save time.
Although one should be careful with this parameter because in certain cases
the error can be decreased after a lot of “motionless” iterations. It's impossible
to automatically detect such cases. We recommend setting 10 iterations, which is
enough for most of problems. For certainty you can set up to 100 iterations.
How much time is required for network training?
The time required for network training depends on the number of
inputs, number of hidden units, amount of data, complexity of the
task and capability of your computer. Complete network training
can continue from several seconds to several hours.
top
|
